Quantcast

Raptor Composite Aircraft

HomeBuiltAirplanes.com

Help Support HomeBuiltAirplanes.com:

jet guy

Well-Known Member
Joined
Aug 15, 2020
Messages
55
Peter has a smaller turbo as the LP!
I'm fairly sure they are the same series of Garrett turbo; I believe 28s, or maybe 31s, although those come with a choice of slightly bigger or smaller compressors and he may have juggled that a bit. But it's pretty much the same compressor map. The turbines will be the same size in the same series.

If you can find some footage of his engine you will see that the hose going from the compressor discharge on that first turbo [HP unit sitting on exhaust manifold] is necked down ridiculously in order to adapt to the much bigger compressor mouth on that aft turbo; the thing looks like a toilet plunger.

You will notice the hot rodders, as in that video I linked to, select turbos where the outlet and corresponding inlet sizes are the SAME diameter. So the first compressor outlet size will match the mouth diameter on the second compressor. This also means that the hot side match will be similar. Since turbos are designed to flow at a particular air flow velocity, matching those sizes is a good clue that they are properly sized.

There is more to designing a staged setup than that of course, but this is a pretty idiot-proof first step. Having so obviously failed that step does not inspire confidence as to what other surprises may be lurking elsewhere in this 'airplane'.
 

pictsidhe

Well-Known Member
Joined
Jul 15, 2014
Messages
8,055
Location
North Carolina
yep.
What he really needs is a certified power plant for hauling a family of five.
A TIO 540 would be a power and reliability upgrade and probably weight reduction too.

If Peter really wants to stick with the Audi conversion, he absolutely needs someone like Ross to do it properly. A few pointers are not going to be enough, especially when he keeps insisting that he is smarter than the people with actual experience.
 

SuperSpinach

Active Member
Joined
Apr 20, 2020
Messages
32
Location
France
Peter responded to jet guy's comment on youtube with this :
(I personnaly don't know much about turbos so I won't be adding anything to the conversation).

Raptor Aircraft
il y a 29 minutes
I took the advice of several different outfits on how to configure the compound turbos. Perhaps no one knew what they were suggesting and it's set up incorrectly. Either way, the hot side housings are identical in size so the order on that side should not really make a difference, they are both moving the same amount of exhaust. On the cold side I'm feeding small into big. 2867 into 2871. At max power the small one boosts about 6psi, the big one about 21psi. Now I don't really want to run much more than the 43psi total pressure that I'm running now. If I inject more fuel I get a lambda below 1 and black smoke. The setup is generating a measured 1000lbs of thrust. The aircraft accelerates well enough to move the mass as quick as most other GA aircraft. I've measured that as well. So, with enough thrust and not overheating and not running the turbine inlet temp over 1500f I'm quite happy that the powerplant is dialed in as much as it can be without a Dyno. I don't need to go to the trouble of putting it on a Dyno just to get a number. Anyway, that's my take on it. Your mileage may vary.
 
Last edited:

lelievre12

Active Member
HBA Supporter
Joined
Jul 15, 2020
Messages
30
Unfortunately, this isn't the case at all; aside from all the known concerns about the redrive and the engine management system, there is a very big and serious problem with that staged turbo setup, which could in fact result in catastrophic engine failure.

Few appear to have noticed this, probably because staged turbos, sometimes called compound, or serial turbos are not used very often even in the automotive world, and especially in light aircraft.

There is a very real risk of overheating of the exhaust valves in that engine, which could result in catastrophic engine failure of course.
I fly a P210 Cessna with larger turbo (than a T210) which provides both engine air and cabin pressuization air. I agree completely with your comments here in that the additional exhaust back pressures demanded on the P210 engine (TSIO520) creates additional EGT which eats exhaust valves and causes exhaust system cracks at a far higher rate than the T210 engine. And at higher altitudes the problem gets worse.

Its simple physics that if the exhaust gases are not expanded, they will remain hotter than if they are.

The OTHER bugaboo is that as the exhaust system is operating 'pressurized' the exhaust valve stem is trying to seal exhaust gases which want to travel upwards into the rocker cover. This forces any oil that is travelling down the stem back out. Exhaust guide wear is therefore much higher as oil cannot make it down the stem to do its job. This is primarily why TSIO520P engines are notorious for eating exhaust valves in as little as 200-400 hours.

The fix on an Avgas engine is to run well ROP so that a significant portion of the unburnt fuel travels into the exhaust system and through evaporation, helps to reduce the EGT's. The other fix is to run a turbo controller which maintains the minimum back pressure necessary to maintain the desired manifold pressure. Absolute pressure controllers = bad. Slope or differential controllers = good.

Of course in a diesel engine, none of this is possible as Lambda values are always 1 or less (LOP not ROP) and required manifold pressures are sky high in order to generate aviation (not tractor) levels of power. So Peter has adopted compound turbos in order to 'solve' the problem perhaps without fully considering the issues of high back pressure and EGT on a stock engine.

So yes, there IS a very real risk of overheating the exhaust valves and the turbos themselves using this much back pressure. However that is not a death sentence in itself, simply an operating limitation that needs to be observed. On my P210, the TIT (turbine inlet temperature) limit is 1650F and on later aircraft with better turbo metallurgy, its 1750F. The corresponding EGT's (which are not limited in the POH) are around 100F less. The TSIO520P exhaust valves (which are made from ni-resist) can survive at these temps, just not for long.

Looking at Pete's data, he is observing the same kind of limits. Since he reprogrammed the ECU to advance and compress the injection cycle (like advancing the ignition on a park ignition engine) he has reduced the full power EGT's to ~1500 which is far less than the turbo limits. Of course at altitude, the TIT will creep up if power is maintained, however as long as Pete reduces power to observe a sensible TIT limit the only ongoing issue will be to borescope the exhaust valves after some time in service to ensure that are not showing signs of overheating. This is done routinely on my plane and so sounds a reasonable approach for Peter, not a showstopper at all.
 
Last edited:

wsimpso1

Super Moderator
Staff member
Log Member
Joined
Oct 18, 2003
Messages
6,984
Location
Saline Michigan
I watched the video again. Not sure how serious that incident was or if his fix was appropriate. Normally, the bolts would be switched to left thread for that, I think. The rotation force might exceed the safety wire ability.
Hmm, we have heard that one in the automotive world too. The flexplate or flywheel is bolted to the rear face of the crank with several bolts, and are standard right hand threads to hold things together while the crank flange and flexplate/flywheel accels/decels right along with the crank. The big deal is to make the clamp loads are large enough that flexplate or flywheel stays put by friction against the firing pulses. The firing pulses look like a sine wave, but is slightly skewed. We had never found justification to work with left hand threads in both car makers I worked for.

The quick view Peter gave us of the bolted joint face looks like the damper hub was scuffed around and may have been slipping. Ideally the bolts produce enough clamp load that the damper hub will not slip on the counterface with some pretty large Factor of Safety. 4 or more is typical. Peter indicated he used lock washers and LocTite to hold these bolts in. I have no grief with LocTite as long as it has high enough temperature resistance, but bolts walking out does make me wonder if he did have enough thermal resistance. Lock washers in this sort of application are another deal - they tend to lose load with time. In automotive use we have long histories without warranty for flat hardened washers and grade 10.9 (yeah metric) bolts torqued way up on flywheel/flexplate attachment.

I hope that his bolts are alloy steel heat treated bolts of SAE grade 8 or better. Softer hex screws in stainless and bright plating are widely available, while the high strength ones may have to be sought out. The more common soft screws are more likely to lose load and allow the damper to slip, with the potential for developing fretting corrosion on the interface and fatigue cracks in the bolts and hub with use.

So, if Peter had soft bolts and they are now hard, and/or if Peter had lockwashers and now has hardened plain ones, he has upgraded and maybe he has put this one to bed. If he replaced in kind, the safety wire will help only a little and this problem may well recur.

Billski
 
Last edited:

pictsidhe

Well-Known Member
Joined
Jul 15, 2014
Messages
8,055
Location
North Carolina
Peter responded to jet guy's comment on youtube with this :
(I personnaly don't know much about turbos so I won't be adding anything to the conversation).

Raptor Aircraft
il y a 29 minutes
I took the advice of several different outfits on how to configure the compound turbos. Perhaps no one knew what they were suggesting and it's set up incorrectly. Either way, the hot side housings are identical in size so the order on that side should not really make a difference, they are both moving the same amount of exhaust. On the cold side I'm feeding small into big. 2867 into 2871. At max power the small one boosts about 6psi, the big one about 21psi. Now I don't really want to run much more than the 43psi total pressure that I'm running now. If I inject more fuel I get a lambda below 1 and black smoke. The setup is generating a measured 1000lbs of thrust. The aircraft accelerates well enough to move the mass as quick as most other GA aircraft. I've measured that as well. So, with enough thrust and not overheating and not running the turbine inlet temp over 1500f I'm quite happy that the powerplant is dialed in as much as it can be without a Dyno. I don't need to go to the trouble of putting it on a Dyno just to get a number. Anyway, that's my take on it. Your mileage may vary.
Well, it must be fine then. It's not like this is the first time Peter has been told about this. I'm sure he's checked it out and pv!=nRT like I thought it did.
 

jet guy

Well-Known Member
Joined
Aug 15, 2020
Messages
55
I fly a P210 Cessna with larger turbo ...This is done routinely on my plane and so sounds a reasonable approach for Peter, not a showstopper at all.
It is in fact a showstopper.

The comparison with a single-turbo Continental engine is not valid, because the problem is with that second turbo, plumbed in series. There are no staged turbos on any recip aircraft engine that I know of.

There is no amount of engine management and fuel delivery fiddling that can alleviate the physical problem of exhaust flow getting bottled up in that way too small second turbo. The cylinder heat will not evacuate the cylinder if there is a downstream logjam. It's as simple as that.

This is a serious problem, especially considering that the life of the test pilot is at stake. Losing engine power on an airplane that needs 90 knots to get unstuck is going to be a very very big problem as you're climbing out on that initial hop.

Whatever other problems the airplane may have, there is no need to have that disaster of a turbo setup that is for sure stressing the exhaust valve. Take that second turbo off and dial in the engine properly, under the supervision of a knowledegable diesel guy.
 
Last edited:

pictsidhe

Well-Known Member
Joined
Jul 15, 2014
Messages
8,055
Location
North Carolina
Hmm, we have heard that one in the automotive world too. The flexplate or flywheel is bolted to the rear face of block with several bolts, and are standard right hand threads to hold things together while the crank flange and flexplate/flywheel accels/decels right along with the crank. The big deal is to make the clamp loads large enough that flexplate or flywheel stays put by friction against the firing pulses. The firing pulses look like a sine wave, but is slightly skewed. We had never found justification to work with left hand threads in both car makers I worked for.

The quick view Peter gave us of the bolted joint face looks like the damper hub was scuffed around and may have been slipping. Ideally the bolts produce enough clamp load that the damper hub will not slip on the counterface with some pretty large Factor of Safety. 4 or more is typical. Peter indicated he used lock washers and LocTite to hold these bolts in. I have no grief with LocTite as long as it has high enough temperature resistance, but bolts walking out does make me wonder if he did have enough thermal resistance. Lock washers in this sort of application are another deal - they tend to lose load with time. In automotive use we have long histories without warranty for flat hardened washers and grade 10.9 (yeah metric) bolts torqued way up on flywheel/flexplate attachment.

I hope that his bolts are alloy steel heat treated bolts of SAE grade 8 or better. Softer hex screws in stainless and bright plating are widely available, while the high strength ones may have to be sought out. The more common soft screws are more likely to lose load and allow the damper to slip, with the potential for developing fretting corrosion on the interface and fatigue cracks in the bolts and hub with use.

So, if Peter had soft bolts and they are now hard, and/or if Peter had lockwashers and now has hardened plain ones, he has upgraded and maybe he has put this one to bed. If he replaced in kind, the safety wire will help only a little and this problem may well recur.

Billski
A single bolt does benefit from appropriate handing. A ring of them does not.

If Peter gave out the dimensions of that joint, there are several people here who could calculate its approximate torque capacity in minutes. He seems VERY averse to having other people check his spitballing, though.
 

BBerson

Light Plane Philosopher
HBA Supporter
Joined
Dec 16, 2007
Messages
13,556
Location
Port Townsend WA
If he replaced in kind, the safety wire will help only a little and this problem may well recur.
Safety wire stretches. I don't think it should be the only method on a torque critical part.
My Grob prop is fitted with big nylock nuts. Yes, the prop turns left hand but still has right hand thread.
Might want a custom bent tab safety plate under those bolt.
 
Last edited:

lelievre12

Active Member
HBA Supporter
Joined
Jul 15, 2020
Messages
30
So I have a 2001 Audi TT Quattro (well, my wife has it - I drive the F350 Diesel Truck) and both of our cars have turbochargers. Nowhere in the manual for either car does it say "don't take your foot off the accelerator too quickly", or "don't leave your foot on the accelerator NEAR idle but not quite there", or "don't stomp on the accelerator". So I'll argue that the engine computer should know what to do in ANY situation of accelerator (or in this case, throttle lever) motion. And if it doesn't, it's not the fault of the operator, but the fault of the programming in the controller/computer. My cars have no issue with whatever I do with the accelerator pedal, stupid though it may be.

I think you've got this backwards. With a mechanical system, you can certainly screw things up by moving levers or pedals too quickly or not quickly enough - Cthulhu knows that's certainly the case with the Airflow Performance fuel injection system on my COZY MKIV's O-360. But with a digital control injection system, one would expect that the lever movement would be telling the controller what the operator wants, and then the controller would do the right thing, whatever that was for the given input - not that it would need to be babied so that it didn't kill the engine because it got confused.
I think its worth looking at this point as it summarises a lot of the naive attitudes in this forum.

The issue is summed as "engine dies = Wasabi gets worried".

Now all agree that if you get in a conventional aircraft (non Fadec) and push ANY lever or button at the wrong time you can damage/destroy your engine in seconds. As one example, if I attempt a takeoff with my TSIO520 and mixture is not full rich I will get detonation and then preignition before I leave the ground and engine will be destroyed soon thereafter.

So we all have a POH which is explicit on what must be done when and all pilots are OK with that.

Now looking at Motec ECU control of an auto diesel, the naive attitude is "oh great, I can drive it like a car!" Wrong. Dead wrong.

The Motec is a simplified racing ECU and mixture works just like manual in the trim tables. So if you set low power and then add load (say like an air-conditioner) the ECU has no way of knowing this and it will maintain that low power level until the engine spools down and dies. There is no magic setting that stops this unless the ECU is told to go to idle control where the mode now changes to provide enough fuel to maintain a set idle speed.

There is no way around this. If you bump up the minimum low power fuel flow in the trim tables to compensate for added accessory loads then the engine will race at low power when accessories are off and never reach idle. Just like it would with a carb or mechanical fuel flow. The only way to overcome is to engage the idle control mode in the ECU.

And so Peter has set his "POH" to say - IDLE SPEED - SET IDLE CONTROL (full back on throttle). Just like any POH would to command a desired response and prevent controls being used incorrectly. Not some other setting that the pilot dreams up (like STUPIDLY pumping the throttle).

Of course the Wasabi guys had no understanding of this and simply thought "drives like a car" just like folks in this forum. Sorry but it ain't so and demeaning Pete's work because of the limitations of our own understanding is nothing better than slander. And when enough mud gets thrown, it sticks.

Now that is not to say that there are not many issues with the project, or that advice from forum members is not valuable, however when untruths are mixed with truths leads to 'poor project' its worth debating this to see just how 'expert' our forum is.

On the point of idle control it is Wasabi = 0, Peter = 1, period. Pete is right and Wasabi should be apologizing in their video to say they were stupid in this regard. Not using their stupidity to demean Peter. Had Wasabi actually followed the POH we'd be discussing more important things like PSRU or flutter.
 
Last edited:

BBerson

Light Plane Philosopher
HBA Supporter
Joined
Dec 16, 2007
Messages
13,556
Location
Port Townsend WA
As previously stated all through this thread, nothing short of turbine power will work for this airframe... assuming it is aerodynamically up to that power.
I don't think there is consensus. A turbine won't go 3000 miles. The mission purpose needs to be redone with less nonsense dreams. I think longer wings could be fitted if the hull is strong enough. And a Continental diesel with real gears.
Or something direct drive.
Need to pick just one primary mission goal.
 

TFF

Well-Known Member
Joined
Apr 28, 2010
Messages
13,444
Location
Memphis, TN
Pretty much Jet Guy is confirming what is buried in this thread when it first got running. Probably 20 pages of discussion. It’s one of those things where we knew Peter read this thread because he reacted to it in one of the videos of past.

If Peter was on board that this is all an experiment in learning, he would rarely get grief. Failure in some element is not a failure if something is learned. The control system is an example. There could be fixes on the board right now with the acceptance that what is on here now is stuck but will be better with the next. What we get is shut up. It’s perfect.

We screamed at the engine; we screamed at the belt drive. He wavered a couple of times and conceded. Only after running out of his own ideas. It’s kind of funny that he gets tons of free advice but can’t filter the good out from the bad. This is all from afar. He could easily had some real face to face solutions from people who know for the price of a cup of coffee. The help just here, and this is not the only place in the world, could have been mined if he was not confrontational about ideas. He would have been closer to 2.0 than .9 Some of these mistakes were resolved at the dawn of aviation. Some were resolved during the composite revolution.

He literally could have peer review of the concepts. He could have easily said here is my engine and gearbox design. Let it churn on the Internet for a month, shuffle the ideas, iterate. You can always lay ground rules. Must be this engine, gearbox must be machined in-house. After about 3-4 iterations it starts coming together way better. And it was free. His audience would be triple because there is a dialogue. Vision gets the ball rolling; Peter has that. Unluckily it’s a pinhole camera instead of a 35 mm if not a 110 Instamatic.
 

TarDevil

Well-Known Member
Joined
Jun 29, 2010
Messages
640
Location
Coastal North Carolina/USA
I don't think there is consensus. A turbine won't go 3000 miles. The mission purpose needs to be redone with less nonsense dreams. I think longer wings could be fitted if the hull is strong enough. And a Continental diesel with real gears.
Or something direct drive.
Need to pick just one primary mission goal.
It won't go 3000 miles, regardless.
 

Toobuilder

Well-Known Member
HBA Supporter
Log Member
Joined
Jan 19, 2010
Messages
4,864
Location
Mojave, Ca
...On the point of idle control it is Wasabi = 0, Peter = 1, period. Pete is right and Wasabi should be apologizing in their video to say they were stupid in this regard. Not using their stupidity to demean Peter. Had Wasabi actually followed the POH we'd be discussing more important things like PSRU or flutter...
I have no idea what the POH says or if the Test guys went off the rails or not, but as the designer, Peter is not washed of responsibility if the process of control is counterintuitive to norms. We would not accept aft stick to command nose down or right stick to command a left roll just because the POH says so, right? The throttle should behave like a throttle, and yes, the ECU should be at least as smart as any "normal" airplane. The SDS system that oversees the 540 in my Rocket certainly is. I have no issues with starting, idle speed, mixture control, throttle bursts... the throttle simply does throttle stuff - no POH needed for anybody who has flown any airplane produced in the last 50 years. If the Raptor ECU is no better than a 60 year old Cessna WRT pilot required management AND its totally dependant on an electrical system that can apparently be depleted in minutes, then Peter has taken a huge step backwards in capability for the aviation masses.
 
Last edited:

TFF

Well-Known Member
Joined
Apr 28, 2010
Messages
13,444
Location
Memphis, TN
Wasabi is presently testing a Stewart 51 with auto engine and gearbox. It’s had problems and with the test program, they test and find issues and the owners come up with solutions for the next time. Test again. It’s taken at least a year to get a dozen flights. It’s probably an Oshkosh winner when the issues are solved.

They are not shoot from the hip. I don’t think Peter has the patience to fix something every flight. Especially if it’s big. He does not want to fix the defects. He wants it flying to get to Oshkosh for the press.

If he can get 3-4 flights that don’t scare the crap out of the pilot, Peter will take over. I don’t think he is ready to push the envelope. Is he ready to see if it acts like a canard or stalls? When the gearbox does it’s documented ten hour part shed, is he cool enough to bring it home? Does he trust the ballistic chute?

3000 miles is LA to Hawaii with reserve. The dream of dreams for a travel only GA pilot.
 

jet guy

Well-Known Member
Joined
Aug 15, 2020
Messages
55
I've tried to post a reply to Peter on his channel, where one of his commenters reposted my remarks on Elliott's channel. But Peter is apparently blocking my comment.

It's interesting to read his reply to that commenter quoting me, which was relayed here as well...

He admits the turbo system may be set up incorrectly, but then proceeds to 'explain' why all is good anyway...

<>'...the hot side housings are identical in size so the order on that side should not really make a difference, they are both moving the same amount of exhaust.'<>

Of course the mass is the same, but the volumetric flow is now much bigger coming out of that first turbo; basic thermodynamics. Having lost maybe half its pressure by driving that first turbine, the flow now needs a flow passage that is close to twice as big.

<>'On the cold side I'm feeding small into big. 2867 into 2871.'<>

Wow. It's worse than I thought. That compressed flow coming out of the first compressor needs to go into a second compressor about half the size, not bigger.

He says he's getting 6 psi on the first turbo and 21 psi on the second, which indicates that the wastegate on that second turbo is NOT opening, as I had assumed. This makes things worse. Fortunately he's not compressing a whole lot on that first turbo, but there is still expansion in that first turbine stage and so the flow is being jammed up in that second turbine.

He claims he is measuring 1,500 F on the EGT and he is 'happy' with that. But how reliable that reading is, is anyone's guess, depending on whether he has properly placed the thermocouple. I would be very surprised if the true EGT was not much higher.

He also says he's happy with the airplane acceleration and the measured thrust of supposedly 1,000 pounds [which would be about 250 hp at most].

<>'I don't need to go to the trouble of putting it on a Dyno just to get a number. Anyway, that's my take on it. Your mileage may vary. 🤔 '<>

Now I have to say here that there is no way I would trust this engine without having its power verified. Running on the dyno is also loading the engine sufficiently to see if it will handle that for an extended run. Simply taxiing the airplane around or doing some ground runs with a prop for a few minutes is not going to do that.

He says he 'consulted' several 'outfits' in putting that turbo system together. And so after some taxiing and ground runs he's good to go. I have to be honest here and say that this is incredibly irresponsible for someone in charge of an airplane project and handing off this airplane to a test pilot. Not to mention unethical.

You have to do whatever it takes to make sure this engine is absolutely flightworthy. If that means getting help from experts and doing a dyno testing program, then how can you just wave all of that away and say, 'your mileage may vary'...?

I just hope no outside pilot is going to be tempted to take this thing into the air. Not with a guy like that in charge.
 
Last edited:

cheapracer

Well-Known Member
Log Member
Joined
Sep 8, 2013
Messages
6,230
Location
Australian
Whatever other problems the airplane may have, there is no need to have that disaster of a turbo setup that is for sure stressing the exhaust valve. Take that second turbo off and dial in the engine properly, under the supervision of a knowledegable diesel guy.
I haven't a clue why he has a compound setup, it's not like it needs low down boost, it's a plane, as you say, just stick a bigger single turbo on.

I actually thought originally he was running a separate 2nd turbo to boost the cabin pressure..


I think its worth looking at this point as it summarises a lot of the naive attitudes in this forum.

Now that is not to say that there are not many issues with the project, or that advice from forum members is not valuable, however when untruths are mixed with truths leads to 'poor project' its worth debating this to see just how 'expert' our forum is.
What you are doing is taking isolated examples and trying for a win of some condescending sorts, even if you were right, and you're not, it's just one example of literally a hundred plus faults throughout the build.

You remind me of a Guy I serviced his rally car during an event many years ago. Service means to carry spare tires and fuel around while he competes through stages, meeting him at the end of each competitive stage called a service point.

I put 'X' amount of fuel calculated just enough to get to the next service point (keep the car light as possible) and he arrives late at that next service point because I hadn't put his fuel cap on the way it needs to be put on, just 2/3rds of a turn. If you put it on like a normal person, turning it all the way 'til it stops, it leaks out on right hand corners, and he ran out of fuel.

Apparently my fault because he was used to it and never told me (besides putting a new darn $5 cap on it!) .... I put his tires and fuel drums out on the ground there and then, and drove home.

Point is, just because Raptor guy is indoctrinated into an obscure way to operate his craft that he's been doing for "70 hours" of engine running, doesn't mean normal people with normal operational expectations, unless told otherwise, are immediately in the wrong. They aren't

And you evidently know little about Motec capabilities.
 
Top